Optimization of a grid-connected renewable energy ...

Optimization of a grid-connected renewable energy system for a case study in Nablus, Palestine

..............................................................................................................................................................

Tamer Khatib1,2 * 1 Department of Electrical, Electronic & System Engineering, Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia, Bangi 43600, Selangor, Malaysia; 2 Palestinian Solar and Sustainable Energy Society, Ramallah, Palestine

.............................................................................................................................................

Abstract

Keywords: PV system; wind system; grid-connected PV system; inverter size; optimization of gridconnected system *Corresponding author: [email protected]

Received 5 December 2012; revised 17 January 2013; accepted 7 February 2013

................................................................................................................................................................................

1 INTRODUCTION According to OSLO accords signed between Israel and the Palestinian authority, the Palestinian energy authority and the Palestinian electricity companies are responsible of the electricity distribution network of the Palestinian authority areas. Meanwhile, the transmission and the generation of the electricity are the responsibilities of the Israeli electricity company/ companies. Based on this, any suggested upgrade/expansion for a distribution power station located in the Palestinian authority area needs a political decision because it cannot be done without cooperation with the Israeli electricity companies. On the other hand, the current capacity of the installed power stations in Nablus is 50 MW, and these stations have not been upgraded/expanded for a long time because of the current political situation. Based on this, the current situation of the distribution network in Nablus is quite critical especially in the peak times of the summer and winter seasons [1– 3].

Therefore, a grid-connected renewable energy system can solve this problem because the installation of such a system does not require any political decision. However, based on the fact that PV systems are clean, environmentally friendly and secure energy sources, PV system installation has played an important role worldwide. However, the drawback of PV system is the high capital cost as compared with conventional energy sources. Therefore, many research works are currently being carried out focusing on optimization of PV systems [4]. Grid-connected PV systems can be divided into two parts: building-integrated PV systems (BiPV) and distribution generation PV (DGPV) systems. BiPV systems usually supply a specific load and inject the excess energy to the grid. GDPV systems inject the whole produced energy to the grid without feeding any local load. The grid-connected systems can consist of a PV array as an energy source only, or another energy source can be in cooperated with the PV array such as wind turbine, diesel system or a storage unit [5].

International Journal of Low-Carbon Technologies 2014, 9, 311– 318 # The Author 2013. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/ by-nc/3.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact [email protected] doi:10.1093/ijlct/ctt007 Advance Access Publication 18 March 2013 311

Downloaded from http://ijlct.oxfordjournals.org/ by guest on October 30, 2015

Recently, the Palestinian energy authority has approved a renewable energy policy to resolve the electricity problem caused by the Palestinian-Israeli energy dispute. Based on this, clear technical aspects in regards to renewable energy systems must enforced. In this research, a renewable energy system consisting of a PV and a wind energy source is proposed to be connected to Nablus city electricity grid. The proposed system is optimally designed taking into consideration maximum system productivity and inverter size. Two evaluating factors are used to optimally choose the configuration of the proposed system namely final yield factor (FY) and capacity factor (CF). As for the inverter, a liner programming optimization is performed to find out the optimum inverter sizing ratio (the rated power of the energy source to the rated power of the inverter). The results show that the use of PV energy sources is more feasible as compared to wind energy sources in Nablus. Therefore, a grid-connected system consisting of PV array only as an energy source is recommended. Moreover, the optimum sizing ratio of the inverter in the proposed system is 1.42.

T. Khatib

2 WEATHER PROFILE FOR THE TARGETED SITE Figure 1 shows daily averages of solar energy for the selected site. The solar energy values are in the range of (0.5–9 kWh/m2). The average solar energy is 5.2 kWh/m2. The highest value is in 312 International Journal of Low-Carbon Technologies 2014, 9, 311– 318

Figure 1. Solar energy profile for the selected site.

Figure 2. Wind speed profile for the selected site.

the middle of the year (June–August) while the lowest value is in the beginning and the end of the year (November–February). According to this, the site has a high solar energy potential, which encourages a PV system choice. Figure 2 shows daily averages of wind speed for the selected site. The wind speed profile fluctuates all over the year with an average value equal to 4.3 m/s. This average value shows that the wind energy potential is slighter higher than the cut in speed of the wind turbine but is still quite acceptable for wind power systems installation. Figure 3 shows the ambient temperature profile for the selected sites. The temperature has its peak value (298C) in the middle of the year (summer) while it is lower in the winter and autumn semesters. However, the average ambient temperature is 18.38C.

3 MODELING OF THE ENERGY SOURCES IN THE PROPOSED SYSTEMS Figure 4 shows the proposed system to be integrated with the electricity grid. The proposed system consists of a PV array and a wind farm, and it is supposed to have a capacity of 30 kWp. This system is supposed to be connected to the 400 V bas bar of a nearby power substation.


However, much research focuses on the PV array size, PV array tilt angle and orientation and inverter size of gird-connected systems. In [6], some parameters have been optimized for an optimum installation of a grid PV system in Spain. These parameters include shading analysis of the system location in order to avoid any external or self shading, wire losses, optimum orientation and tilt angel, inverter size. However, no size optimization for the PV array is mentioned. Moreover, the inverter size is selected by intuitive consideration. In [7], size optimization of a grid-connected system is presented. The variables that are optimized in this research are the PV generator type, inverter type and PV array tilt angle. Mathematical models for PV array, inverter and solar radiation on tilt surface are developed. Using these models and real solar radiation and temperature records, a possible design space is obtained using four types of PV generator, three types of inverter and seven tilt angle values. Finally, the configuration that investigates the maximum efficiency is selected as an optimum choice. In [8], the impact of PV array area, orientation and tilt angle, inverter sizing ratio on the feasibility of grid-connected system is highlighted. A grid-connected system mathematical model as well as real weather data are used in an iterative simulation to optimally size/select these variables. In [9], a PV grid-connected systems optimization is presented. The main objective of this research is to find the optimal number and type of the PV modules and the DC/AC converters, the PV modules optimal tilt angle, the optimal arrangement of the PV modules within the available installation area and the optimal distribution of the PV modules among the DC/AC converters. This optimization is done using by modeling the system using real weather data by an iterative simulation. AI techniques were also employed for optimizing grid-connected PV systems, whereas in [10] particle swarm optimization is used in multiobjective optimization problem for optimally design of photovoltaic grid-connected systems. The optimization’s decision variables are the optimal number of the PV modules, the PV modules optimal tilt angle, the optimal placement of the PV modules within the available installation area and the optimal distribution of the PV modules among the DC/AC converters. The design optimization aims towards the maximization of both the economic and environmental benefits received by the use of grid-connected system. The main objective of this article is to optimally design a grid-connected renewable energy system for a selected site in Nablus. This is investigated by finding out the optimum configuration of the energy sources as well as the inverter size in the proposed system.

Optimization of a grid-connected renewable energy system

The energy produced by a PV module can be calculated in terms of the solar energy and the ambient temperature as follows [4]: EPV ðtÞ ¼ APV ESUN ðtÞhPV ðtÞhINV hWIRE

ð1Þ

where APV, ESUN, hPV, hINV and hWIRE are the areas of the PV array, the solar energy, the PV module conversion efficiency, the inverter conversion efficiency and the wire efficiency, respectively. However, the effect of temperature on the conversion efficiency of a PV module can be expressed by [4],

hPV ðtÞ ¼ hPVREF ½1 bðTc ðtÞ Tcref Þ

ð2Þ

where hPVRef, b, Tc and Tcref are the reference PV module conversion efficiency, temperature coefficient for the efficiency, cell temperature and reference cell temperature, respectively. The cell temperature is calculated using the ambient temperature as follows by: Tc ðtÞ Tambient ¼

NOCT 20 GðtÞ 800

ð3Þ

The coefficients (C1 – C4) can be calculated by a fitting tool such as MATLAB fitting tool.

4 OPTIMIZATION OF THE ENERGY SOURCES IN THE PROPOSED SYSTEM Figure 3. Ambient temperature profile for the selected site.

In this section, the optimization of the system is divided into two parts, namely energy sources optimization and inverter

Figure 4. Description of the proposed system.

International Journal of Low-Carbon Technologies 2014, 9, 311– 318 313


where Tambient, Ttest and G are the ambient temperature, nominal operation cell temperature and the solar radiation, respectively. The output power of a wind generator which is given as a function of wind speed is illustrated in Figure 5. From Figure 5, the characteristic of a wind turbine can be described as follows: 8 9 < 0; if! Vcutout , V , Vcutin = if ð4Þ P ; ! V ¼ VRATED : RATED c2 V ; if c4 V P ¼ c1 e þ c3 e ; ! Vcutin , V , VRATED

T. Khatib

size optimization. The proposed system is supposed to be a hybrid PV/wind system. However, many configurations consisting of a PV array and a wind farm can give a 30-kWp system. Moreover, the system could consist of PV array only or wind farm only. Therefore, an evaluation of the possible configuration must be done in order to find out the optimum choice. In this research, the evaluation of the possible configuration is done using two factors, namely final yield factor and capacity factor. The final yield is defined as the annual, monthly or daily net AC energy output of the system divided by the peak power of the installed PV array at standard test conditions and it is given by [11], FY ¼

EPV ðkWh=yearÞ PVWP ðkWpÞ

ð5Þ

The capacity factor (CF) is defined as the ratio of the actual annual energy output to the amount of energy the PV array would generate if it operated at fully rated power (Pr) for 24 h per day for a year [11], CF ¼

FY EPV ¼ 8760 PR 8760

ð6Þ

where EPV is the annual energy generated by a PV array and PR is the rated power of a PV array. However, a design space for the desired system is generated using a linear programming. This design space contains the possible configurations of PV array and wind farm that give a 30 kWp. After that, each possible configuration is evaluated using the factors mentioned in 314 International Journal of Low-Carbon Technologies 2014, 9, 311– 318

Equations (5 and 7) and hourly solar energy, wind speed and ambient temperature data. Finally, the configuration that has the highest values of these factors is chosen as an optimum choice.

4.1 Optimization of the DC/AC inverter size for the PV array The rated power of a PV array must be optimally matched with inverter’s rated power in order to achieve maximum PV array output power [12]. The optimal inverter sizing depends on local solar radiation and ambient temperature and inverter performance [13, 14]. For instance, under low solar radiation levels, a PV array generates power at only part of its rated power and consequently the inverter operates under part load conditions with lower system efficiency. PV array efficiency is also affected adversely when an inverter’s rated capacity is much lower than the PV rated capacity. On the other hand, under overloading condition, excess PV output power that is greater than the inverter rated capacity is lost [15, 16]. This is to say that optimal sizing of PV inverter plays a significant role in increasing PV system efficiency and feasibility. The efficiency of an inverter is calculated by:

hðtÞ

Pin ðtÞ PLoss ðtÞ Pin ðtÞ

ð7Þ

where Pin(t) and Ploss(t) are the instantaneous input power and power loss during the conversion. Ignoring the DC – DC converter efficiency, the input power to the PV system is the output power of the PV module. The


Figure 5. A wind turbine characteristic.


PLoss is not constant but depends on many conditions, which makes it difficult to be calculated. Thus, an alternative model for inverter efficiency needs to be developed in order to estimate the inverter’s output power. Figure 6 shows an efficiency curve for a commercial inverter obtained from the datasheet. The curve describes the inverter’s efficiency in terms of input power and inverter rated power. The efficiency curve can be described by a power function as follows: 8 9 PPV PPV > if > > C2 þ C3 ; ! . 0> < h ¼ c1 = PINVR PINVR ð8Þ P > > > > : h ¼ 0; if! PV ¼ 0 ; PINVR

RS ¼

PPVRATED PINVRATED

ð9Þ

where PPVRATED is the rated power of the PV array and PINVRATED is the rated power of the inverter. The objective function of the optimization problem is maximizing the annual average inverter efficiency which is formulated in terms of the daily averages of solar radiation (G), ambient temperature (T) and inverter rated power (P) and is given by, P1 Figure 6. Typical efficiency curve for an inverter.

MAX : hannual ¼

hDaily ¼ 366

366

P1

266

PPVinput =PINVR 366

ð10Þ

Figure 7. Efficiency curves for the three chosen inverters.



where PPV and PVINVc are PV module output power and inverter’s rated power respectively while C1 – C3 are the model coefficients.

A MATLAB fitting tool can be used for calculating the developed inverter model coefficients, C1 – C3. Therefore, samples of the inverter’s efficiency curves shown in Figure 6 must be taken for the purpose of curve fitting using the MATLAB fitting tool. An intensive number of samples must be taken for a specific part of the curve as in zone B, a less intensive number of samples are taken from zone A, while few samples are taken from zone C, as described in Figure 6. In this work, three commercial inverters of rated powers, 5, 50 and 100 kW are considered. The reason for choosing these inverters is to match the three types of loads, namely, low (5 kW), medium (50 kW) and high (100 KW). Figure 7 shows the efficiency curves which have been constructed based on the model described in Equation 8 for the chosen inverters types. Using the MATLAB fitting tool, the developed models coefficients are calculated as shown in Table 1. The optimum size of an inverter is represented by the ratio, Rs which is the PV array rated power to the inverter rated power, and it can be descried mathematically as follows:

T. Khatib

Table 1. Inverter models coefficients

5 kW 50 kW 100 kW Average

Table 3. Proposed PV system cost

C1

C2

C3

Item

Unit price (USD)

Quantity

Price

Life time

20.2418 20.5879 20.3253 20.385

21.127 21.105 21.143 21.125

96.10 97.76 97.49 97.12

PV array Support structure Inverter Circuit breakers Fuses Transmission lines Civil work Maintains Land Total

3.7/Wp 50 12 000 1500 400 – 1000 300/year 5/m2

30 kWp 30 1 2 2 1 km – – 200 m2

111,000 1500 12,000 3000 800 1500 1000 6000 1000 137,800.00

20 – 20 20 –

Table 2. Examples of systems’ possible configuration evaluation Wind farm size (kWp)

FY (MWh/kWp.year)

CF (%)

0 5.0000 10.0000 15.0000 20.0000 25.0000 30.0000

30.0000 25.0000 20.0000 15.0000 10.0000 5.0000 0

1.2557 1.3664 1.4771 1.5878 1.6985 1.8092 1.9199

14.3349 17.0166 20.2346 24.1678 29.0843 35.4055 43.8337

The optimization problem is then done by using the efficiency curve based optimization, which is an iterative method. The optimization process starts by obtaining the PV system specifications such as PV array rated power, temperature coefficient and maximum power point tracker efficiency. In addition, the hourly solar energy and ambient temperature for the targeted area must be obtained in order to calculate the PV array output power. A set of Rs values is used in the iterative loop, the rated capacity of the inverter is calculated after defining the value of Rs and then the PV array output power is calculated using PV array model. Here, the developed inverter models are used to estimate the efficiency hour by hour through a specific period of time, and then the annual efficiency is calculated and stored in an array. This loop will be repeated iteratively until reaching the maximum value of Rs then a search for the maximum efficiency value and its index (optimum Rs) is conducted. However, this process is done for the adopted sites considering low, medium and high loads (5, 50 and 100 kW inverters) [17 – 19].

5 RESULTS AND DISCUSSION Table 2 shows a part of the evaluated configurations. From the table, it is clear that the system that consists of a 30 kWp (without wind energy source) is the most productive configuration among the other possible configuration whereas its FY and CF are 1.7 MWh/kWp.year and 29.1%, respectively. Meanwhile the choice that consists only of wind energy source is the less productive system as compared with the other configuration, whereas the FY and CF of this choice are 1.256 MWh/kWp.year and 14.3%, respectively. In addition, the

316 International Journal of Low-Carbon Technologies 2014, 9, 311– 318

Table 4. Comparison between PV system and Wind system for the selected site

Annual energy produced (MWh) System capital cost Cost of Energy (USD/kWh) FY (MWh/kWp.year) CF(energy produced/energy rated) (%) Payback period (years)

PV system

Wind system

57.6 137,800.00 0.1196 1.92 43.8 17.1

37 68,000.00 0.096 1.23 14.3 12.4

obtained results show that increasing the PV source share in the proposed system increases the productivity of the system. Based on this, a grid-connected PV system consisting of 30 kWp is recommended for the targeted site. Table 3 shows the proposed system cost. The cost of the proposed system is $137,800.00 while the cost of the energy that produced by the system is $0.1196, which is cheaper than the energy bought from the Israeli distribution grid ($0.14). However, such a PV system has a payback period of about 17 years. To ensure the feasibility of the chosen system, a brief comparison is conducted with a 30-kWp wind power system, as detailed in Table 4. From the table, the capital cost of the wind power system is less than the capital cost of the PV system by 50%. Moreover, the cost of the energy that is produced by the wind power system is slightly lower than the one that is produced by the PV system. Furthermore, the energy production of the PV system is almost twice the energy production of the wind power system and this can be realized by the high FY factor and the capacity factor of the PV system when compared with the wind power system. As for the inverter size, Figure 8 shows the searching for the optimum size considering the solar energy and ambient temperature profile of the selected sites. The figure shows the optimum sizing ration of the invert in the proposed PV system is 1.42, which means that a 21-kWp inverter is recommended for the proposed system.


PV array size (kWp)

– 20 – 20


6 CONCLUSION In this research, a 30-kWp PV system was proposed to be connected to the electricity grid in Nablus, Palestine. The system’s energy productivity was evaluated using two factors: final yield factor and capacity factor. Moreover, a size optimization for the inverter in the proposed system is conducted using a linear programming optimization. The results showed that the final yield factor and the capacity factor of the proposed systems are 1.92-MWh/kWp.year, 43.8%, respectively, while these factors for a 30-kWp wind power system located at the same site are 1.23 MWh/kWp.year and 14.3%. In addition, the optimum sizing ratio of the inverter in the proposed system is 1.42 with maximum efficiency equals to 96.2%.

REFERENCES [1] Khaitb T. Design of photovoltaic water pumping systems at minimum cost for Palestine. J Appl Sci 2010;10:2773 – 84. [2] Mahmoud M, Ibrik I. Field experience on solar electric power systems and their potential in Palestine. Renew Sustain Energy Rev 2003;7:531– 43. [3] Ibrik I, Mahmoud M. Energy efficiency improvement procedures and audit results of electrical, thermal and solar applications in Palestine. Energy Policy 2005;33:651– 8. [4] Khatib T, Mohamed A, Sopian K, et al. Optimal sizing of building integrated hybrid PV/diesel generator system for zero load rejection for Malaysia. Energy Buildings 2011;43:3430 – 35.

[5] Eltawil M, Zhao Z. Grid-connected photovoltaic power systems: technical and potential problems: a review. Renew Sustain Energy Rev 2010;14: 112 – 29. [6] Fernandez-Infantesa A, Contrerasa J, Bernal-Agustın J. Design of grid connected PV systems considering electrical, economical and environmental aspects: a practical case. Renew Energy 2006;31:2042– 62. [7] Notton G, Lazarov G, Stoyanov L. Optimal sizing of a grid-connected PV system for various PV module technologies and inclinations, inverter efficiency characteristics and locations. Renew Energy 2010;35:541– 54. [8] Deb Mondol J, Yohanis Y, Norton B. Comparison of measured and predicted long term performance of grid a connected photovoltaic system. Energy Convers Manage 2007;48:1065– 80. [9] Koutroulis K. Methodology for the design optimisation and the economic analysis of grid-connected photovoltaic systems. IET Renew. Power Gener 2009;3:476– 92. [10] Kornelakis A. Multiobjective particle swarm optimization for the optimal design of photovoltaic grid-connected systems. Sol Energy 2010;84: 2022–33. [11] Kymakis E, Kalykakis S, Papazoglou T. Performance analysis of a grid connected photovoltaic park on the island of Crete. Energy Conv Manage 2009;50:433 –8. [12] Decker B, Jahn U, Rindelhardt U, et al. The German 1000-roofphotovoltaic-programme: system design and energy balance. In: 11th European Photovoltaic Solar Energy Conference, Montreux, Switzerland, 1992, 1497 – 500. [13] Macagnan MH, Lorenzo E. On the optimal size of inverters for grid connected PV systems. In: 11th European Photovoltaic Solar Energy Conference, Montreux, Switzerland, 1992, 1167 –70. [14] Jantsch M, Schmidt H, Schmid J. Results of the concerted action on power conditioning and control. In: 11th Photovoltaic Solar Energy Conference, Montreux, Switzerland, 1992, 1589 – 1593. [15] Keller L, Affolter P. Optimizing the panel area of a photovoltaic system in relation to the static inverter-practical results. Sol Energy 1995;55:1– 7.



Figure 8. Searching for the optimum sizing ratio of the inverter.

T. Khatib

[16] Coppye W, Maranda W, Nir Y, et al. Detailed comparison of the inverter operation of two grid-connected PV demonstration systems in Belgium. In: 13th European Photovoltaic Solar Energy Conference, Nice, France, 1995, 1881–4. [17] Khatib T, Mohamed A, Sopian K. Optimization of a PV/wind micro-grid for rural housing electrification using a hybrid iterative/genetic algorithm: case study of Kuala Terengganu, Malaysia. Energy Buildings 2012;47:321–31.

[18] Omar A, Shaari S. Sizing verification of photovoltaic array and gridconnected inverter ratio for the Malaysian building integrated photovoltaic project Int. J Low-Carbon Tech 2009;4:254 – 57. [19] Aldali Y, Henderson D, Muneer T. A 50 MW very large-scale photovoltaic power plant for Al-Kufra, Libya: energetic, economic and environmental impact analysis. Int J Low-Carbon Tech 2011;6:277 –93.


318 International Journal of Low-Carbon Technologies 2014, 9, 311– 318